宜配屋

一，mnist数据集

形如上图的数字手写体就是mnist数据集。

二，GAN原理(生成对抗网络)

GAN网络一共由两部分组成：一个是伪造器(Generator，简称G)，一个是判别器(Discrimniator，简称D)

一开始，G由服从某几个分布(如高斯分布)的噪音组成，生成的图片不断送给D判断是否正确，直到G生成的图片连D都判断以为是真的。D每一轮除了看过G生成的假图片以外，还要见数据集中的真图片，以前者和后者得到的损失函数值为依据更新D网络中的权值。因此G和D都在不停地更新权值。以下图为例：

在v1时的G只不过是一堆噪声，见过数据集(real images)的D肯定能判断出G所生成的是假的。当然G也能知道D判断它是假的这个结果，因此G就会更新权值，到v2的时候，G就能生成更逼真的图片来让D判断，当然在v2时D也是会先看一次真图片，再去判断G所生成的图片。以此类推，不断循环就是GAN的思想。

三，训练代码

import argparse
import os
import numpy as np
import math
 
import torchvision.transforms as transforms
from torchvision.utils import save_image
 
from torch.utils.data import DataLoader
from torchvision import datasets
from torch.autograd import Variable
 
import torch.nn as nn
import torch.nn.functional as F
import torch
 
os.makedirs("images", exist_ok=True)
 
parser = argparse.ArgumentParser()
parser.add_argument("--n_epochs", type=int, default=200, help="number of epochs of training")
parser.add_argument("--batch_size", type=int, default=64, help="size of the batches")
parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate")
parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient")
parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient")
parser.add_argument("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
parser.add_argument("--latent_dim", type=int, default=100, help="dimensionality of the latent space")
parser.add_argument("--img_size", type=int, default=28, help="size of each image dimension")
parser.add_argument("--channels", type=int, default=1, help="number of image channels")
parser.add_argument("--sample_interval", type=int, default=400, help="interval betwen image samples")
opt = parser.parse_args()
print(opt)
 
img_shape = (opt.channels, opt.img_size, opt.img_size) # 确定图片输入的格式为(1，28，28)，由于mnist数据集是灰度图所以通道为1
cuda = True if torch.cuda.is_available() else False
 
 
class Generator(nn.Module):
 def __init__(self):
  super(Generator, self).__init__()
 
  def block(in_feat, out_feat, normalize=True):
   layers = [nn.Linear(in_feat, out_feat)]
   if normalize:
    layers.append(nn.BatchNorm1d(out_feat, 0.8))
   layers.append(nn.LeakyReLU(0.2, inplace=True))
   return layers
 
  self.model = nn.Sequential(
   *block(opt.latent_dim, 128, normalize=False),
   *block(128, 256),
   *block(256, 512),
   *block(512, 1024),
   nn.Linear(1024, int(np.prod(img_shape))),
   nn.Tanh()
  )
 
 def forward(self, z):
  img = self.model(z)
  img = img.view(img.size(0), *img_shape)
  return img
 
 
class Discriminator(nn.Module):
 def __init__(self):
  super(Discriminator, self).__init__()
 
  self.model = nn.Sequential(
   nn.Linear(int(np.prod(img_shape)), 512),
   nn.LeakyReLU(0.2, inplace=True),
   nn.Linear(512, 256),
   nn.LeakyReLU(0.2, inplace=True),
   nn.Linear(256, 1),
   nn.Sigmoid(),
  )
 
 def forward(self, img):
  img_flat = img.view(img.size(0), -1)
  validity = self.model(img_flat)
  return validity
 
 
# Loss function
adversarial_loss = torch.nn.BCELoss()
 
# Initialize generator and discriminator
generator = Generator()
discriminator = Discriminator()
 
if cuda:
 generator.cuda()
 discriminator.cuda()
 adversarial_loss.cuda()
 
# Configure data loader
os.makedirs("../../data/mnist", exist_ok=True)
dataloader = torch.utils.data.DataLoader(
 datasets.MNIST(
  "../../data/mnist",
  train=True,
  download=True,
  transform=transforms.Compose(
   [transforms.Resize(opt.img_size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5])]
  ),
 ),
 batch_size=opt.batch_size,
 shuffle=True,
)
 
# Optimizers
optimizer_G = torch.optim.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
optimizer_D = torch.optim.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
 
Tensor = torch.cuda.FloatTensor if cuda else torch.FloatTensor
 
# ----------
# Training
# ----------
if __name__ == '__main__':
 for epoch in range(opt.n_epochs):
  for i, (imgs, _) in enumerate(dataloader):
   # print(imgs.shape)
   # Adversarial ground truths
   valid = Variable(Tensor(imgs.size(0), 1).fill_(1.0), requires_grad=False) # 全1
   fake = Variable(Tensor(imgs.size(0), 1).fill_(0.0), requires_grad=False) # 全0
   # Configure input
   real_imgs = Variable(imgs.type(Tensor))
 
   # -----------------
   # Train Generator
   # -----------------
 
   optimizer_G.zero_grad() # 清空G网络 上一个batch的梯度
 
   # Sample noise as generator input
   z = Variable(Tensor(np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim)))) # 生成的噪音，均值为0方差为1维度为(64，100)的噪音
   # Generate a batch of images
   gen_imgs = generator(z)
   # Loss measures generator's ability to fool the discriminator
   g_loss = adversarial_loss(discriminator(gen_imgs), valid)
 
   g_loss.backward() # g_loss用于更新G网络的权值，g_loss于D网络的判断结果 有关
   optimizer_G.step()
 
   # ---------------------
   # Train Discriminator
   # ---------------------
 
   optimizer_D.zero_grad() # 清空D网络 上一个batch的梯度
   # Measure discriminator's ability to classify real from generated samples
   real_loss = adversarial_loss(discriminator(real_imgs), valid)
   fake_loss = adversarial_loss(discriminator(gen_imgs.detach()), fake)
   d_loss = (real_loss + fake_loss) / 2
 
   d_loss.backward() # d_loss用于更新D网络的权值
   optimizer_D.step()
 
   print(
    "[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f]"
    % (epoch, opt.n_epochs, i, len(dataloader), d_loss.item(), g_loss.item())
   )
 
   batches_done = epoch * len(dataloader) + i
   if batches_done % opt.sample_interval == 0:
    save_image(gen_imgs.data[:25], "images/%d.png" % batches_done, nrow=5, normalize=True) # 保存一个batchsize中的25张
   if (epoch+1) %2 ==0:
    print('save..')
    torch.save(generator,'g%d.pth' % epoch)
    torch.save(discriminator,'d%d.pth' % epoch)

运行结果：

一开始时，G生成的全是杂音：

然后逐渐呈现数字的雏形：

最后一次生成的结果：

四，测试代码：

导入最后保存生成器的模型：

from gan import Generator,Discriminator
import torch
import matplotlib.pyplot as plt
from torch.autograd import Variable
import numpy as np
from torchvision.utils import save_image
 
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
Tensor = torch.cuda.FloatTensor
g = torch.load('g199.pth') #导入生成器Generator模型
#d = torch.load('d.pth')
g = g.to(device)
#d = d.to(device)
 
z = Variable(Tensor(np.random.normal(0, 1, (64, 100)))) #输入的噪音
gen_imgs =g(z) #生产图片
save_image(gen_imgs.data[:25], "images.png" , nrow=5, normalize=True)

生成结果：